This proposal describes the development of new methods for quantum chemistry (both SCF and electron correlation methodologies), molecular dynamics, and protein folding simulations. In all cases the codes have been highly optimized and demonstrate superior performance when compared to plausible alternatives; both serial and massively parallel implementations will be pursued. Applications include understanding charge separation in the photosynthetic reaction center, determination of protein structure from a minimal set of NMR data, and studies of helix stability via quantum chemical calculations in solution.